hypocretin A receptor in the basolateral amygdala

Psychoneuroendocrinology (2014) 40, 17—26

Available online at www.sciencedirect.com

ScienceDirect j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n

Anxiolytic function of the orexin 2/hypocretin A receptor in the basolateral amygdala David H. Arendt a,b,c,e, James Hassell b, Hao Li b, Justin K. Achua b,c,e, Douglas J. Guarnieri g, Ralph J. DiLeone g, Patrick J. Ronan c,d,e,f, Cliff H. Summers b,c,* a

Department of Psychiatry, Indiana University of School of Medicine, Indianapolis, IN 46202, USA Department of Biology, University of South Dakota, Vermillion, SD 57069, USA c Neuroscience Group, Division of Basic Biomedical Sciences, Sanford School of Medicine, University of South Dakota, Vermillion, SD 57069, USA d Avera Research Institute, Sioux Falls, SD, USA e Research Service, Sioux Falls VA Medical Center, Sioux Falls, SD, USA f Department of Psychiatry, Sanford School of Medicine, University of South Dakota, Vermillion, SD, USA g Division of Molecular Psychiatry, Ribicoff Research Facilities, Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA b

Received 14 August 2013; received in revised form 1 October 2013; accepted 23 October 2013

KEYWORDS Orexin; Hypocretin; Anxiety; Amygdala; Orexin 2 receptor; shRNA; Knockdown; Basolateral amygdala

Summary The orexin/hypocretin system interacts with many of the same circuitries contributing to stress-associated disorders like depression and anxiety. These include potentially reciprocal connections with corticotropin releasing factor (CRF) neurons which drive the hypothalamic—pituitary—adrenal (HPA) endocrine response in addition to having an anxiogenic effect in the central amygdala (CeA). Antagonism of the orexin type 1 receptor (Orx1) in the hypothalamus has also been shown to block panic attacks. However, few studies have investigated the effect of orexinergic signaling in the basolateral amygdala (BLA) which is responsible for contextual fear, and modulates the activity of the CeA. To this end, we chronically stressed c57bl/6 mice with social defeat and examined the gene expression of the orexin receptors in the BLA. We found that the transcripts for the Orx1 and Orx2 receptors diverged in the BLA with Orx1 increasing and Orx2 decreasing in animals that were susceptible to the chronic defeat. These changes were not seen in the prelimbic cortex (PrL) which sends efferents to the BLA. We then

* Corresponding author at: Department of Biology, University of South Dakota, 414 East Clark Street, Vermillion, SD 57069-2390, USA. Tel.: +1 605 677 6177; fax: +1 605 677 6557. E-mail addresses: [email protected], [email protected] (C.H. Summers). 0306-4530/$ — see front matter # 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.psyneuen.2013.10.010

18

D.H. Arendt et al. tried to recapitulate these expression patterns in the BLA using short hairpin interfering sequences delivered by adeno-associated viruses to knock down the orexin receptors. While the Orx1 knockdown did reduce locomotor activity, it did not decrease depressive or anxious behaviors. Knocking down the Orx2 receptors in the BLA increased anxious behavior as measured by reduced social preference and reduced time spent in the center of an open field. Due to the divergent expression patterns of the two receptors in response to chronic stress, orexinergic activity in the BLA may be responsible for bidirectional modulation of anxious behavior. Furthermore, these data raise the possibility that an Orx2 agonist may serve as an effective means to treat anxiety disorders. # 2013 Elsevier Ltd. All rights reserved.

1. Introduction The orexin (hypocretin) system is involved in regulating motivated behavior and arousal (de Lecea et al., 1998; Deadwyler et al., 2007; Lin et al., 1999; Sakurai et al., 1998). The neural systems associated with these behaviors are also often affected by disorders such as anxiety and depression (Krishnan and Nestler, 2010; Pittenger and Duman, 2008; Ressler and Mayberg, 2007; Yehuda et al., 1996). Our work and that of others has shown that orexin may play a role in depression due to its involvement in states such as chronic stress and learned helplessness which appear to be part of the etiology of the disorder (Arendt et al., 2013; Lutter et al., 2008; Nollet et al., 2011; Nollet and Leman, 2013; Scott et al., 2011). In addition, orexin is involved in anxiety disorders which are highly comorbid with depression (Avolio et al., 2011; Johnson et al., 2012a,b; Kessler et al., 2003; Li et al., 2010). Most of this work examines the activity of the two heterotrimeric G protein coupled receptors of the orexin system, type 1 and type 2 (Orx1, Orx2; de Lecea et al., 1998; Sakurai et al., 1998). Two endogenous ligands bind to these receptors, orexin A and B (OrxA, OrxB). While the Orx2 receptor binds both forms with equal affinity, the Orx1 receptor binds OrxA with a tenfold greater affinity relative to OrxB (Sakurai et al., 1998). The central (CeA) and basolateral (BLA) subnuclei of the amygdala contain orexin projections and express the transcript for orexin receptors (Marcus et al., 2001; Peyron et al., 1998). There may also be cross talk between the two systems as amygdalar subregions project to orexin neurons (Sakurai et al., 2005). Application of either OrxA or OrxB excites CeA neurons (Bisetti et al., 2006) and produces increased behavioral anxiety (Avolio et al., 2011). The BLA is involved in fear/anxiety as well (Barot et al., 2009; Orsini and Maren, 2012; Tye et al., 2011). Inhibition of glutamatergic signaling in the BLA blocks anxiety responses as measured by social preference (Sajdyk and Shekhar, 1997). Furthermore, upregulating CREB in the BLA, a transcription factor positively associated with LTP and memory formation (Kida, 2012), increases behavioral anxiety (Wallace et al., 2004). In addition to its role in anxiety, the BLA processes contextual information about fear stimuli which is then conveyed to the CeA (Barot et al., 2009; Orsini and Maren, 2012). Despite the BLA’s role in these behaviors, little is known about orexinergic signaling in this area and its impact on anxiety. In the past, we found a correlation between higher expression of learned helplessness, as measured by the forced swim test, and amounts of OrxA and the Orx1 gene transcript in the

amygdala (Arendt et al., 2013). In the study reported herein we examine the effect of chronic social defeat on the gene expression of Orx1 and Orx2 receptors. Our hypothesis, that anxiety and depression are mediated through activation of the Orx1 receptor in the BLA, was tested by determining if its knockdown would produce anxiolytic and antidepressive effects. We further hypothesize that the Orx2 receptor in the BLA functionally counter-balances the Orx1 receptor, with anxiolytic and antidepressive effects. We hypothesize that the knockdown of the Orx2 receptor will be anxiogenic and depressive. We tested these using adeno-associated viruses to insert stably expressing short hairpin sequences to knockdown the mRNA transcripts for each of the orexin receptors.

2. Materials and methods 2.1. Animals Adult male c57bl/6 mice (Harlan, Indianapolis; N = 59) were housed in pairs during the first half of the study prior to the social defeat (N = 28), and housed singly in the second half (N = 22). A separate cohort of animals was used to validate the viral vectors (N = 9). Mice were on a 12:12 reversed light— dark cycle (lights off at 9 AM) with food and water provided ad libitum. Retired CD1 breeder mice were used for social defeat and the social preference tests. All testing took place between 10 AM and 2 PM. All experiments were executed in a manner that minimized suffering and the number of animals used, in accordance with the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23), and approved by the Institutional Animal Care and Use Committee of the University of South Dakota.

2.2. Behavioral measures/experimental design The first half of this study examines the effects of chronic social stress on the expression of Orx1 and Orx2 receptor transcripts in the BLA and PrL. After ten days of chronic social defeat, an animal’s susceptibility or resilience to social defeat was assessed on day eleven by a social preference test (Fig. 1). Orexin receptor expression in susceptible and resilient animals in the appropriate brain regions was examined using qPCR. The second half of the study was carried out in a separate cohort of mice, for which orexin receptors were chronically knocked down in the BLA of unstressed animals. After orexin

Anxiolytic function of BLA Orx2 receptors

19

Figure 1 (A) Sequence and timing of social defeats (by a novel CD1 mouse once/day for 10 days) and following social preference test (day 11). (B) Social preference scores based on the amount of time test mice spent near and away from a social target delineate animals that were resilient and susceptible to social defeat. ‘‘Resilient’’ animals were defeated, but still showed a social preference (>100% time spent with social target/time spent away from the social target  100) similar to ‘‘Controls’’. Defeated animals that displayed a social aversion (<100%) relative to controls and were defined as ‘‘Susceptible’’.

receptor knockdown, behavioral endpoints of depression and anxiety were assessed using measures intended to quantify behavior associated with depression and anxiety, listed below (Fig. 4C). Testing took place between 10 AM and 2 PM under red light unless otherwise listed. 2.2.1. 10 Day social defeat Cages were divided longitudinally by perforated Plexiglas to allow olfactory and minimal tactile contact. The CD1 mice were housed singly on one side of this enclosure for one week to establish territorial attachment to the home cage. For the following ten days social defeat was established in c57bl/6 test mice by placing one on the side of the divider where the territorial CD1 resided, after which aggression rapidly ensued. The c57bl/6 was left with the CD1 for up to 5 min, or until 5 attacks on the intruder were observed. Once removed, the c57bl/6 was placed on the opposite side of the divider where it remained in visual and olfactory (but not physical) contact with the CD1 for the next 24 h. This sequence took place for 10 days, each day with a novel CD1; no additional defeat followed on day 11 when the social preference test was administered. To control for social housing over the 10 days, two c57bl/6 mice were housed on opposite sides of a similar divider, but not allowed to fight. 2.2.2. Social preference Testing of social avoidance or preference took place in a square box (40 cm3) with a transparent perforated cylinder (diameter 12 cm) set against the middle of one wall. This test was used for both socially defeated and AAV-treated Orx receptor knockdown mice. After ten days of chronic social defeat, on day eleven c57bl/6 mice were moved directly into the social preference test from the opponent’s home cage. In AAV-treated mice, with injection on day one, the social preference test followed on day 22. The c57bl/6 mouse was allowed to explore the box with the empty cylinder for 2.5 min before being removed, and then the empty cylinder was exchanged for one containing a non-aggressive CD1 mouse. Test animals were not defeated on the day of the

social preference test (days 11 or 22). The c57bl/6 was then put back into the box and allowed to explore the cylinder with a social target for an additional 2.5 min. All interactions were monitored with Ethovision software (Noldus; Leesburg VA) to measure the amount of time the c57bl/6 mouse spent in close proximity (3 cm) to the cylinder, as well as the total distance (cm) traveled in the apparatus. The test animal’s final social preference score was determined by using the proportion of time the c57bl/6 mouse spent in close proximity (3 cm) to the cylinder when it was occupied vs when it was empty ([time within close proximity when occupied]/ [time within close proximity when empty]  100 = social preference score). Typically, undefeated control mice spent more time in close proximity to the social target giving them a social preference score of greater than 100%. A portion of the defeated mice displayed social preference greater than 100% and were thus termed ‘‘resilient’’, as these preference scores were similar to control animals. In contrast, on the day after defeat ‘‘susceptible’’ mice spent less time in close proximity to the social target producing a social preference score less than 100%. 2.2.3. Sucrose preference test This test quantifies an animal’s preference for a naturally rewarding substance, a sweet sucrose solution, and was carried out over 6 days beginning on day 15 after AAV injection. On the first and second days (15 and 16 post-AAV) animals were presented with 2 bottles of water to acclimate them to a choice of bottles. One of the water bottles was replaced with a sucrose solution (1%) for the third through the sixth days (17—20 post-AAV). Sucrose and water bottles were alternated each day to prevent the influence of location bias. The third day (17 post-AAV) was not recorded, but served as a means to acclimate the mice to the taste of sucrose. Weight of bottles containing sucrose solution and water were recorded on the fourth-sixth days (18—20 post-AAV) to determine the percentage of fluid intake that was sucrose solution ([sucrose volume]\[sucrose + water volume]  100 = % sucrose intake).

20 2.2.4. Open field test This test measures an animal’s exploratory behavior of a novel environment, and was administered on day 21 after AAV injection. Animals are less likely to explore the center of the apparatus if they are more anxious. Mice were placed in a square box (40 cm  40 cm  40 cm) for 5 min. Trials were recorded and examined using Ethovision software (Noldus, Leesburg, VA) for the amount of time animals spent in the center (28 cm) and outer walls (6 cm) of the box as well as the total distance traveled. 2.2.5. Elevated plus maze Similar to the open field, this test was administered on day 21 after AAV injection, and measures the amount of time an animal spends exploring an exposed area vs one that is more protected. Animals were placed on the center of the Elevated Plus Maze (EPM) facing the same closed arm for every trial. Movement was recorded using Ethovision software (Noldus, Leesburg, VA) for the total amount of time the animal spent in the open and closed arms and total distance traveled. This test was performed on the same day as the open field test with a 5 min interval separating the two tests. 2.2.6. Forced swim test The forced swim test was administered on day 23 after AAV injection, and measures immobility in an inescapable cylinder of water, which is correlated with learned helplessness. Exposure to stressful stimuli apriori increases immobility times while antidepressants decrease immobility. Mice were placed in a 1 L cylinder of 25 8C water and recorded for 8 min. The last 5 min of the video were scored for the total amount of time that the animals spent in an immobile state. Immobility was defined as either no limb movement, or the minimal amount of limb movement required to keep the animal’s nose above the water.

2.3. qPCR Defeated mice were sedated with 5% isoflurane and decapitated with brains frozen on dry ice. The majority of each brain region of interest was micro-dissected from frozen coronal sections (200 mm) on a Peltier cold plate (T = 25 8C; Physitemp, Clifton, NJ) using a blunt 24 gauge needle. Prelimbic cortex was taken from approximately 1.54—1.98 mm anterior to bregma while the BLA was sampled from 1.22 to 2.06 posterior to bregma. Regions were immediately injected into lysis buffer for subsequent RNA purification (Life Technologies, Grand Island, NY). Final RNA samples were quantified using a Nanodrop ND-1000 spectrophotometer (Thermo Scientific, Waltham, MA). Purified RNA was used for complementary DNA (cDNA) synthesis in 20 ml reactions using the High Capacity cDNA Archive Kit (Life Technologies, Grand Island, NY). The cDNA product was used in the qPCR reactions on a Step One Plus Real-Time PCR System (Life Technologies, Grand Island, NY). Samples were examined using Taqman Assay On Demand primer/probe sets (Life Technologies, Grand Island, NY) for Orx1 (mm 01185776_m1) and Orx2 (mm 01179312_m1). Each sample was run in duplicate and normalized to the expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH; mm03302249q1). The qPCR was performed at 50 8C for

D.H. Arendt et al. 2 min and 95 8C for 10 min, followed by 45 cycles at 95 8C for 15 s and 60 8C for 1 min. Changes in gene expression were analyzed using the 2 DDCT method (Livak and Schmittgen, 2001), comparing all samples to the average DCT value of the control animals. Values were expressed as mean fold change  standard error of the mean (SEM).

2.4. AAV shRNA injections We used a similar protocol as Hommel et al. (2003) for the viral vectors designed to knock down the expression of the orexin receptors. These AAV vectors were generated by the DiLeone lab. Mice were sedated with 5% isoflurane before being placed in a Kopf stereotaxic frame. During the surgery, sedation was maintained with 3% isoflurane. Holes were drilled into the skull and bilateral injections were made targeting the BLA (AP 1.5, ML 3.25, DV 4.75) according to bregma. Viral delivery was done manually with a stereotaxically mounted Hamilton syringe over ten minutes. The syringe was left in place for an additional 5 min to minimize reflux migration of the viral vector up the injection track. We injected 400 nL of a serotype II adeno-associated virus which delivered a plasmid that coded for shRNA knockdown sequences and an EGFP reporter. Behavioral testing took place 18 days after viral injections to allow expression of the shRNA product. Separate hairpin RNA structures were designed to target specific regions of the Orx1 and Orx2 mRNAs. Critical to this structure was a 24-nucleotide sequence that had a consensus sequence on each orexin receptor mRNA. Each designed shRNA contains a 24-nucleotide antisense sequence followed by the sense sequence (matched to a consensus sequence on each orexin receptor mRNA), with sense and antisense sequences separated by a loop sequence (CTTCCTGTCA). Three sequences were used for experiments with sense target sequences: Orx1: CCAAAGGTCCCCACAGACATATTC Orx2: AGAAACCCTTCAGTGGGACTTAAC Scrambled control: CGGAATTTAGAAACCCGGCTCCAC The oligonucleotides had SapI and XbaI overhangs to allow for ligation downstream of the mU6pro region of a modified pAAV-MCS vector, pAAV-shRNA. This vector was designed to coexpress hairpin RNAs, under the control of a mouse U6 promoter and an SV40 polyadenylation site, as well as EGFP controlled by an independent CMV promoter and hGH polyadenylation sequence (Hommel et al., 2003; Sharf et al., 2010).

2.5. Statistics Group comparisons of different brain regions were performed using one way ANOVA, followed by post hoc comparisons of control, susceptible, and resilient groups using the Newman— Keuls test. Correlative effects between social preference and gene effects were performed with regression analysis. Following AAV injections, animals were included in the final analysis if they unilaterally or bilaterally expressed GFP in the BLA as behavioral results were not statistically different (scramble: F = 7.6, p > 0.1; Orx1: F = 2.9, p > 0.16; Orx2: F = 2.2, p > 0.2) for left, right or bilaterally accurate injections. All

Anxiolytic function of BLA Orx2 receptors

21

differences for AAV treated animals were compared using ttests between animals that received the control virus and those receiving the Orx1 or Orx2 viral treatment.

3. Results 3.1. Chronic defeat produces susceptible and resilient populations Mice that underwent the 10 day defeat protocol had a wider range of social preference times relative to controls that were not defeated (Fig. 1). Defeated animals were divided according to whether they exhibited social preference (‘‘resilient’’, social preference score > 100%) or social avoidance (‘‘susceptible’’, social preference score < 100%; Krishnan et al., 2007). All but one of the control mice not previously exposed to an aggressive CD1 had social preference scores greater than 100%. Normally, rodents display a social preference for novel conspecifics. However, it may be possible that social preference scores in control animals were also influenced by the novelty of the strain of mouse (CD1) used to assess affiliation or avoidance during the test.

3.2. Chronic defeat decreases Orx2 receptor mRNA in the BLA There were no significant differences in orexin receptor gene expression between control and resilient groups for either of

A

2.5

10 Day Defeat - BLA

*#

3.3. Viral shRNA knockdown of Orx2 in the BLA induces anxious behavior The shRNA constructs reduced the amount of endogenous mRNA to 40.2% of control expression for the Orx1 sequence (t7 = 3.458, p < 0.011; Fig. 4A) and 51.9% for the Orx2 sequence (t5 = 2.852, p < 0.036; Fig. 4A). As behavioral outcomes were comparable, animals with both unilateral and bilateral delivery of shRNA to the basolateral complex were included in the final behavioral analysis (Fig. 4B). Neither of the AAV shRNA constructs targeted at the Orx1 or Orx2 receptor mRNA sequences significantly affected behavioral

B

1.5

1.0

1.4

10 Day Defeat - BLA

1.2

Orx2 (Gene Expression)

Orx1 (Gene Expression)

2.0

the brain areas (BLA, PrL) we examined. In the BLA, susceptible animals showed an increase in Orx1 transcription (F 2,13 = 7.753, p < 0.006; Fig. 2A) and a decrease for Orx2 (F 2,11 = 7.142, p < 0.010; Fig. 2B). These changes were not seen in the PrL cortex for either the Orx1 (F 2,25 = 0.764, p < 0.476; Fig. 2C) or Orx2 (F 2,25 = 0.556, p < 0.581; Fig. 2D) transcripts. Changes in BLA Orx receptor gene expression also translated to significant correlations between social preference score and expression of Orx1 (F 1,14 = 24.322, p < 0.001, r2 = 0.635; negative regression; Fig. 3A) and Orx2 (F 1,12 = 23.365, p < 0.001, r2 = 0.661; positive regression; Fig. 3B) receptor transcripts. These relationships were not significant for Orx1 (F 1,26 = 0.021, p < 0.887, r2 < 0.001; Fig. 3C) or Orx2 (F 1,26 = 0.001, p < 0.981, r2 < 0.001; Fig. 3D) mRNA in the PrL.

1.0

*#

0.8 0.6 0.4

0.5 0.2

0.0

0.0

Control

C

2.5

Resilient

10 Day Defeat - Prelimbic

Susceptible

Cortex

Control

D

10 Day Defeat - Prelimbic

Susceptible

Cortex

1.8 1.6

Orx2 (Gene Expression)

2.0

Orx1 (Gene Expression)

2.0

Resilient

1.5

1.0

1.4 1.2 1.0 0.8 0.6 0.4

0.5

0.2 0.0

0.0

Control

Resilient

Susceptible

Control

Resilient

Susceptible

Figure 2 In the basolateral amygdala (BLA), (A) susceptible animals had significantly elevated amounts of Orx1 mRNA relative to control (*) and resilient (#) animals. (B) This effect in susceptible animals occurred with a concomitant decrease in BLA Orx2 mRNA. In the prelimbic cortex (PrL), there were no differences between the groups for the (C) Orx1 or (D) Orx2 receptor transcripts. *,#p < 0.05.

22

D.H. Arendt et al.

A 2.2

B

10 Day Defeat - BLA p<0.001 r2=0.635

1.6 1.4 1.2 1.0

Control Resilient Susceptible

0.8

Control Resilient Susceptible

1.0 0.9 0.8 0.7

p<0.001 r2=0.661

0.6

0.6

0.5

0

C

10 Day Defeat - BLA

1.2 1.1

1.8

Orx2 (Gene Expression)

Orx1 (Gene Expression)

2.0

50

100

200

0

Social Preference (%)

D

Cortex

10 Day Defeat - Prelimbic

2.0 1.8

Orx 2 (Gene Expression)

Orx 1 (Gene Expression)

150

[time social target] / [time no target]

1.6 1.4 1.2 1.0 0.8 0.6 Control Resilient Susceptible

200

Cortex

Control Resilient Susceptible

1.8

0.2

100

[time social target] / [time no target]

2.0

0.4

50

Social Preference (%)

10 Day Defeat - Prelimbic

2.2

150

1.6 1.4 1.2 1.0 0.8 0.6 0.4

0.0 0

50

100

150

200

0

50

100

150

Social Preference (%)

Social Preference (%)

[time social target] / [time no target]

[time social target] / [time no target]

200

Figure 3 Social preference scores were significantly correlated with BLA expression of (A) Orx1 and (B) Orx2 mRNA. In the BLA there was a significant ( p < 0.001, r2 = 0.635) negative regression between the Orx1 mRNA and social preference, and a significant ( p < 0.001, r2 = 0.661) positive regression between the Orx2 mRNA and social preference. These correlations were absent in the PrL for (C) Orx1 and (D) Orx2 receptor transcripts.

measures of depression for anhedonia in the sucrose preference test (Orx1, t11 = 0.736, p < 0.477; Orx2, t12 = 0.382, p < 0.709; Fig. 5A), or immobility in the forced swim test (Orx1, t10 = 0.529, p < 0.608; Orx2, t11 = 0.738, p < 0.476; Fig. 5B). With regard to anxious behavior, knocking down orexin receptor expression in the BLA did not have any effect on EPM performance for either Orx1 (t12 = 0.159 p < 0.877) or Orx2 (t12 = 0.835, p < 0.420; Fig. 6A). However, knocking down Orx2 receptor expression, but not Orx1, significantly decreased social preference (Orx2, t11 = 2.373, p < 0.037; Orx1, t10 = 0.898 p < 0.390, Fig. 6B). Similarly, the Orx2 receptor AAV knockdown had an anxiogenic effect, decreasing the amount of time spent in the center of the arena of the open field test compared to controls (t11 = 2.704, p < 0.021; Fig. 6C). The same effect in the open field was not seen for the Orx1 receptor (t12 = 1.014, p < 0.331). Total locomotor activity in the social preference test was significantly reduced for the Orx1 AAV treatment group (t10 = 2.468, p < 0.033) but not for mice receiving the Orx2 treatment (t12 = 1.510, p < 0.159; Fig. 6B). There were no differences in open field locomotor activity between the control and

experimental groups (Orx1, t12 = 0.242, p < 0.813; Orx2, t11 = 1.541, p < 0.152; Fig. 6C).

4. Discussion This study provides evidence that Orx2 receptor signaling in the BLA constrains anxious responses in mice, which may include general and social anxiety. The results also suggest that a more inclusive orexin-derived response may ensue following social defeat, since Orx1 and Orx2 receptor gene expression is changed in susceptible mice. What is more, the gene expression in orexin receptor subtypes appears to be regulated in opposite directions following social defeat, suggesting opposing functional responses of Orx1 and Orx2 receptors. While both the BLA and PrL play a role in anxiety/ fear learning (Orsini and Maren, 2012; Sajdyk and Shekhar, 1997; Vidal-Gonzalez et al., 2006), chronic social defeat only changed mRNA expression of orexin receptors in the BLA. A decrease in Orx2 mRNA was seen in susceptible animals, and localized BLA Orx2 receptor knockdown led to the expression of anxiety-like behaviors in social preference and open field

Anxiolytic function of BLA Orx2 receptors

23

Figure 4 Viral shRNA constructs significantly (*) reduced the expression of the endogenous mRNA sequence for the respective orexin receptor. (A) Gene expression comparisons were made between the control group that received a nonspecific ‘‘scramble’’ virus and the experimental group which received the ‘‘knockdown’’ virus targeting the Orx1 or Orx2 receptor. (B) Behavioral data were analyzed for animals with expression of GFP (*) in the basolateral complex. (C) Sequence and timing of a series of tests commonly used to detect depression and anxiety related behaviors to which transfected animals were exposed. *p < 0.05.

tests in defeat-naı¨ve mice. Taken together, these results suggest that Orx2 receptors in the BLA are functionally associated with alleviating anxiety. After 10 days of defeat, clearly distinctive social preference scores varied across our experimental population, producing three easily definable groups based on treatment and social response (control, resilient, and susceptible; Fig. 1) similar to that of a study by Krishnan and colleagues who used a similar protocol in a much larger study (Krishnan et al., 2007). While the prelimbic region (PrL) of the prefrontal cortex is a potential site for orexin to regulate anxious behavior since this region is associated with anxiety states (Miller et al., 2012; Wall et al., 2012), we did not see any changes in Orx1 or Orx 2 receptor transcription in this area. In the BLA, susceptible animals expressed a divergent pattern of orexin receptor transcript where Orx1 was increased, and Orx2 was decreased (Fig. 2). What is more, regressions comparing Orx1 and Orx2 gene expression in the BLA with social preference scores strongly suggest that the action of each of the two receptors within the BLA are functionally opposing that of the other with respect to social anxiety (Fig. 3). This pattern of opposing receptor function is similar to another study that examined knockout mice in the forced swim test where Orx1 null animals displayed less learned helplessness while Orx2 knockouts exhibited more (Scott et al., 2011). While our study employed a different experimental protocol, the contrasting effects of the orexin

receptors in both studies reinforce the possibility that the two receptors could ultimately have different, if not opposite, influences on stress-related disorders. Knocking down the Orx1 transcript in the BLA did not appear to have a significant effect in any of the behaviors we observed that are indicative of depression or anxiety, although the treatment did appear to inhibit locomotion during the social preference test (Fig. 6B). Altering locomotor activity via orexin signaling is not entirely without precedent; as orexin knockout mice display reduced movement during resident intruder testing (Kayaba et al., 2003). In contrast, injecting OrxA or OrxB into the CeA can increase locomotor activity (Avolio et al., 2011). Furthermore, the locomotor change is not surprising as the amygdala is part of a larger circuitry responsible for defensive flight responses (Canteras et al., 2010; Tannure et al., 2009). Since chronic defeat increased the Orx1 gene transcript, we expected that knocking it down in healthy animals would have antidepressive or anxiolytic effects. Others have shown that icv injections of OrxA, which preferentially binds to the Orx1 receptor, increases anxious behavior (Suzuki et al., 2005). Furthermore, Johnson and colleagues have repeatedly shown that systemic antagonism of the Orx1 receptor (SB334867) has a panicolytic effect in rats (Johnson et al., 2010, 2012b). Potentially, the differential results obtained in socially defeated and defeat-naı¨ve mice suggest that the additional stress associated with social defeat may activate a

24

D.H. Arendt et al.

A

Sucrose Preference

60 58 56 54 52 50 48

Control

B

260

Orx1 KD

Orx 2 KD

Social Preference (%)

220

200

180

Elevated Plus Maze

60

40

20

0

Control

B

Forced Swim

240

Immobility (seconds)

80

Time on Open Arms (seconds)

62 (as % total fluid)

Sucrose Consumption

64

400

Orx1 KD

Orx2 KD

Social Preference

2500

*

300

2000

1500

200

*

1000

Locomotion (cm)

66

[time social target] / [time no target]

A

100 500

160 0

Orx 2 KD

Figure 5 Knockdown of either orexin receptor in the BLA produced no significant differences in behavioral measures of depression for either the (A) sucrose preference test or (B) forced swim test relative to the control treatment.

more complete orexin receptor response in the basolateral amygdala. More experimentation is necessary to discern whether Orx1 receptors in BLA have the potential to counteract the anxiolytic responses mediated by BLA Orx2 receptors. The AAV shRNA targeted destruction of the Orx2 transcript in the BLA had an anxiogenic effect (Fig. 6). Specifically, this group of animals exhibited decreased preference for a novel CD1 mouse, and spent less time in the center of an open field. While we did not see a significant effect on the EPM, this may be expected as a previous study that systemically blocked both Orx1 and Orx2 receptors also did not appear to alter EPM performance (Steiner et al., 2012). Similarly, systemic delivery of an Orx1 antagonist, SB-334867, did not affect anxiety behavior on the EPM (Rodgers et al., 2013). Despite the lack of anxiety-like behavior on the EPM, our results suggest that Orx2 receptor activity in the BLA is associated with reduction of both general and social anxiety. This effect with the Orx2 receptor knockdown complements the decrease of Orx2 transcript from chronic defeat, given that both are associated with decreased social preference. However, the combination of ten day social defeat and AAV knockdown of Orx2 receptor (and potentially Orx1) may increase our knowledge of the relationship between anxiety and depression, which are highly comorbid disorders, both of which may be influenced by orexin (Arendt et al., 2013; Kessler et al., 2005; Nollet and Leman, 2013). Ultimately, the two parts of this experiment make a compelling argument for the Orx2 receptor having an anxiolytic function in the BLA.

0

Control

C

250

Orx1 KD

Orx2 KD

Open Field 2500

200 2000

150 1500

*

100

50

1000

Locomotion (cm)

Orx1 KD

Time in Center (seconds)

Control

500

0

0

Control

Orx1 KD

Orx2 KD

Figure 6 Knocking down the Orx2 receptor in the BLA increased anxiety-like behaviors. (A) While none of the treatment groups exhibited altered EPM performance, (B) the viral shRNA Orx2 knockdown significantly decreased (*) social preference relative to controls. Locomotion was unaffected in the social preference test with Orx2 knockdown. There was no significant change in social preference for animals that received the Orx1 knockdown despite a significant decrease in the total amount of locomotor activity. (C) Orx2 knockdown also decreased time spent in the center of an open field apparatus. There were no significant changes in locomotion due to Orx1 or Orx2 knockdown in the open field. *p < 0.05.

While we conclude that Orx2 receptors in the BLA have anxiolytic properties, we are unable to distinguish the exact mechanism at this point. This is because neither the microdissection technique used on the chronic defeat animals, nor the AAV shRNA vectors distinguishes between neuronal subtypes in the BLA (pyramidal projection neurons vs interneurons). The medial nucleus of the CeA (mCeA) is the main

Anxiolytic function of BLA Orx2 receptors output projection of the amygdala and its activation increases anxious behavior (Krettek and Price, 1978; Tye et al., 2011). The lateral nucleus of the central amygdala (lCeA) and medial paracapsular intercalated cells (ITC) both forward GABAergic projections to the mCeA effectively inhibiting anxiety (Avolio et al., 2011; Orsini and Maren, 2012). The BLA projects to all three of these areas amplifying the number of potential mechanisms that could explain why Orx2 receptor signaling would ultimately inhibit anxiety as our data suggest. Orexin system receptors, Orx1 and Orx2, are associated with Gq and Gs proteins, with previous work implicating a primarily stimulatory role for both, with the caveat that Gi association has also been noted (Akbari et al., 2011; Bernard et al., 2003; de Lecea et al., 1998; Hoang et al., 2003; Karteris et al., 2005; Kukkonen, 2013; Sakurai et al., 1998). If the Orx2 receptor has stimulatory actions in the BLA pyramidal neurons (Bisetti et al., 2006), then those neurons would have to make connections with either the lCeA or ITCs to ultimately inhibit the anxiety producing activities of the mCeA. Alternatively, if we assume stimulatory Orx2 receptors reside on BLA GABAergic interneurons, inhibition of BLA pyramidal neurons that directly project to the mCeA should limit amygdalar anxiogenic output (Avolio et al., 2011). These two proposed circuitries would be switched in the event that the Orx2 receptor is coupled instead to a Gi protein, such that Orx2 Gi receptors in the BLA inhibited GABA interneurons which would disinhibit pyramidal output to ITC and/or lCeA. The alternative is for Orx2 Gi output to directly inhibit BLA pyramidal neurons projecting to mCeA. If there is an anxiogenic role for Orx1 receptors to influence the BLA after stress, as our preliminary evidence suggests, any hypothetical circuitry would have to balance the two receptor outputs.

Summary In conclusion, Orx2 receptors in the BLA have the potential to alleviate general and social anxieties. There is also a possibility that the two receptors associated with this system may functionally counteract each other, providing a mechanism to bi-directionally control anxiety via Orx1 and Orx2 receptors in the BLA. An Orx2 receptor specific agonist may provide a means to subvert the complex interconnectivity of the amygdala and act as a novel treatment for anxiety disorders. The efficacy of intranasal OrxA application in rats and primates (Deadwyler et al., 2007; Dhuria et al., 2009a,b) raises the possibility that non-invasive therapeutic activation of orexin receptors may be applicable for use in human clinical trials and eventual treatments for general and social anxiety.

Conflict of interest statement None declared.

Acknowledgements We would like to thank Dr. Diane Lagace and Mirela Hasu for their invaluable advice regarding the social preference test. This work was supported by NIH RR15567, and by anonymous donors to the Summers’ lab via the USD Foundation. The contents of this paper do not represent the views of the Department of Veterans Affairs or the United States Government

25

References Akbari, E., Motamedi, F., Davoodi, F.G., Noorbakhshnia, M., Ghanbarian, E., 2011. Orexin-1 receptor mediates long-term potentiation in the dentate gyrus area of freely moving rats. Behav. Brain Res. 216, 375—380. Arendt, D.H., Ronan, P.J., Oliver, K.D., Callahan, L.B., Summers, T.R., Summers, C.H., 2013. Depressive behavior and activation of the orexin/hypocretin system. Behav. Neurosci. 127, 86—94. Avolio, E., Alo, R., Carelli, A., Canonaco, M., 2011. Amygdalar orexinergic-GABAergic interactions regulate anxiety behaviors of the Syrian golden hamster. Behav. Brain Res. 218, 288—295. Barot, S.K., Chung, A., Kim, J.J., Bernstein, I.L., 2009. Functional imaging of stimulus convergence in amygdalar neurons during Pavlovian fear conditioning. PLoS ONE 4, e6156. Bernard, R., Lydic, R., Baghdoyan, H.A., 2003. Hypocretin-1 causes G protein activation and increases ACh release in rat pons. Eur. J. Neurosci. 18, 1775—1785. Bisetti, A., Cvetkovic, V., Serafin, M., Bayer, L., Machard, D., Jones, B.E., Muhlethaler, M., 2006. Excitatory action of hypocretin/ orexin on neurons of the central medial amygdala. Neuroscience 142, 999—1004. Canteras, N.S., Resstel, L.B., Bertoglio, L.J., Carobrez Ade, P., Guimaraes, F.S., 2010. Neuroanatomy of anxiety. Curr. Top. Behav. Neurosci. 2, 77—96. de Lecea, L., Kilduff, T.S., Peyron, C., Gao, X., Foye, P.E., Danielson, P.E., Fukuhara, C., Battenberg, E.L., Gautvik, V.T., Bartlett 2nd, F.S., Frankel, W.N., van den Pol, A.N., Bloom, F.E., Gautvik, K.M., Sutcliffe, J.G., 1998. The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc. Natl. Acad. Sci. U.S.A. 95, 322—327. Deadwyler, S.A., Porrino, L., Siegel, J.M., Hampson, R.E., 2007. Systemic and nasal delivery of orexin-A (hypocretin-1) reduces the effects of sleep deprivation on cognitive performance in nonhuman primates. J. Neurosci. 27, 14239—14247. Dhuria, S.V., Hanson, L.R., Frey 2nd, W.H., 2009a. Intranasal drug targeting of hypocretin-1 (orexin-A) to the central nervous system. J. Pharm. Sci. 98, 2501—2515. Dhuria, S.V., Hanson, L.R., Frey 2nd, W.H., 2009b. Novel vasoconstrictor formulation to enhance intranasal targeting of neuropeptide therapeutics to the central nervous system. J. Pharmacol. Exp. Ther. 328, 312—320. Hoang, Q.V., Bajic, D., Yanagisawa, M., Nakajima, S., Nakajima, Y., 2003. Effects of orexin (hypocretin) on GIRK channels. J. Neurophysiol. 90, 693—702. Hommel, J.D., Sears, R.M., Georgescu, D., Simmons, D.L., DiLeone, R.J., 2003. Local gene knockdown in the brain using viral-mediated RNA interference. Nat. Med. 9, 1539—1544. Johnson, P., Molosh, A., Truitt, W., Shekhar, A., 2012a. Orexin, stress and anxiety/panic states. Prog. Brain Res. 198, 133—161. Johnson, P.L., Samuels, B.C., Fitz, S.D., Lightman, S.L., Lowry, C.A., Shekhar, A., 2012b. Activation of the orexin 1 receptor is a critical component of CO2-mediated anxiety and hypertension but not bradycardia. Neuropsychopharmacology 37, 1911—1922. Johnson, P.L., Truitt, W., Fitz, S.D., Minick, P.E., Dietrich, A., Sanghani, S., Traskman-Bendz, L., Goddard, A.W., Brundin, L., Shekhar, A., 2010. A key role for orexin in panic anxiety. Nat. Med. 16, 111—115. Karteris, E., Machado, R.J., Chen, J., Zervou, S., Hillhouse, E.W., Randeva, H.S., 2005. Food deprivation differentially modulates orexin receptor expression and signaling in rat hypothalamus and adrenal cortex. Am. J. Physiol. Endocrinol. Metab. 288, E1089— E1100. Kayaba, Y., Nakamura, A., Kasuya, Y., Ohuchi, T., Yanagisawa, M., Komuro, I., Fukuda, Y., Kuwaki, T., 2003. Attenuated defense response and low basal blood pressure in orexin knockout mice. Am. J. Physiol. Regul. Integr. Comp. Physiol. 285, R581—R593.

26 Kessler, R.C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K.R., Rush, A.J., Walters, E.E., Wang, P.S., 2003. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). J. Am. Med. Assoc. 289, 3095—3105. Kessler, R.C., Berglund, P., Demler, O., Jin, R., Merikangas, K.R., Walters, E.E., 2005. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 62, 593—602. Kida, S., 2012. A functional role for CREB as a positive regulator of memory formation and LTP. Exp. Neurobiol. 21, 136—140. Krettek, J.E., Price, J.L., 1978. A description of the amygdaloid complex in the rat and cat with observations on intra-amygdaloid axonal connections. J. Comp. Neurol. 178, 255—280. Krishnan, V., Han, M.H., Graham, D.L., Berton, O., Renthal, W., Russo, S.J., Laplant, Q., Graham, A., Lutter, M., Lagace, D.C., Ghose, S., Reister, R., Tannous, P., Green, T.A., Neve, R.L., Chakravarty, S., Kumar, A., Eisch, A.J., Self, D.W., Lee, F.S., Tamminga, C.A., Cooper, D.C., Gershenfeld, H.K., Nestler, E.J., 2007. Molecular adaptations underlying susceptibility and resistance to social defeat in brain reward regions. Cell 131, 391—404. Krishnan, V., Nestler, E.J., 2010. Linking molecules to mood: new insight into the biology of depression. Am. J. Psychiatry 167, 1305—1320. Kukkonen, J.P., 2013. Physiology of the orexinergic/hypocretinergic system: a revisit in 2012. Am. J. Physiol. Cell Physiol. 304, C2— C32. Li, Y., Li, S., Wei, C., Wang, H., Sui, N., Kirouac, G.J., 2010. Orexins in the paraventricular nucleus of the thalamus mediate anxiety-like responses in rats. Psychopharmacology (Berl) 212, 251—265. Lin, L., Faraco, J., Li, R., Kadotani, H., Rogers, W., Lin, X., Qiu, X., de Jong, P.J., Nishino, S., Mignot, E., 1999. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98, 365—376. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2( DDC(T)) method. Methods 25, 402—408. Lutter, M., Krishnan, V., Russo, S.J., Jung, S., McClung, C.A., Nestler, E.J., 2008. Orexin signaling mediates the antidepressant-like effect of calorie restriction. J. Neurosci. 28, 3071—3075. Marcus, J.N., Aschkenasi, C.J., Lee, C.E., Chemelli, R.M., Saper, C.B., Yanagisawa, M., Elmquist, J.K., 2001. Differential expression of orexin receptors 1 and 2 in the rat brain. J. Comp. Neurol. 435, 6—25. Miller, M.M., Morrison, J.H., McEwen, B.S., 2012. Basal anxiety-like behavior predicts differences in dendritic morphology in the medial prefrontal cortex in two strains of rats. Behav. Brain Res. 229, 280—288. Nollet, M., Gaillard, P., Minier, F., Tanti, A., Belzung, C., Leman, S., 2011. Activation of orexin neurons in dorsomedial/perifornical hypothalamus and antidepressant reversal in a rodent model of depression. Neuropharmacology 61, 336—346. Nollet, M., Leman, S., 2013. Role of orexin in the pathophysiology of depression: potential for pharmacological intervention. CNS Drugs 27, 411—422. Orsini, C.A., Maren, S., 2012. Neural and cellular mechanisms of fear and extinction memory formation. Neurosci. Biobehav. Rev. 36, 1773—1802. Peyron, C., Tighe, D.K., van den Pol, A.N., de Lecea, L., Heller, H.C., Sutcliffe, J.G., Kilduff, T.S., 1998. Neurons containing hypocretin (orexin) project to multiple neuronal systems. J. Neurosci. 18, 9996—10015.

D.H. Arendt et al. Pittenger, C., Duman, R.S., 2008. Stress, depression, and neuroplasticity: a convergence of mechanisms. Neuropsychopharmacology 33, 88—109. Ressler, K.J., Mayberg, H.S., 2007. Targeting abnormal neural circuits in mood and anxiety disorders: from the laboratory to the clinic. Nat. Neurosci. 10, 1116—1124. Rodgers, R.J., Wright, F.L., Snow, N.F., Taylor, L.J., 2013. Orexin-1 receptor antagonism fails to reduce anxiety-like behaviour in either plus-maze-naive or plus-maze-experienced mice. Behav. Brain Res. 243C, 213—219. Sajdyk, T.J., Shekhar, A., 1997. Excitatory amino acid receptors in the basolateral amygdala regulate anxiety responses in the social interaction test. Brain Res. 764, 262—264. Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R.M., Tanaka, H., Williams, S.C., Richarson, J.A., Kozlowski, G.P., Wilson, S., Arch, J.R., Buckingham, R.E., Haynes, A.C., Carr, S.A., Annan, R.S., McNulty, D.E., Liu, W.S., Terrett, J.A., Elshourbagy, N.A., Bergsma, D.J., Yanagisawa, M., 1998. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92 1 page following 696. Sakurai, T., Nagata, R., Yamanaka, A., Kawamura, H., Tsujino, N., Muraki, Y., Kageyama, H., Kunita, S., Takahashi, S., Goto, K., Koyama, Y., Shioda, S., Yanagisawa, M., 2005. Input of orexin/ hypocretin neurons revealed by a genetically encoded tracer in mice. Neuron 46, 297—308. Scott, M.M., Marcus, J.N., Pettersen, A., Birnbaum, S.G., Mochizuki, T., Scammell, T.E., Nestler, E.J., Elmquist, J.K., Lutter, M., 2011. Hcrtr1 and 2 signaling differentially regulates depression-like behaviors. Behav. Brain Res. 222, 289—294. Sharf, R., Sarhan, M., Brayton, C.E., Guarnieri, D.J., Taylor, J.R., DiLeone, R.J., 2010. Orexin signaling via the orexin 1 receptor mediates operant responding for food reinforcement. Biol. Psychiatry 67, 753—760. Steiner, M.A., Lecourt, H., Jenck, F., 2012. The brain orexin system and almorexant in fear-conditioned startle reactions in the rat. Psychopharmacology (Berl) 223, 465—475. Suzuki, M., Beuckmann, C.T., Shikata, K., Ogura, H., Sawai, T., 2005. Orexin-A (hypocretin-1) is possibly involved in generation of anxiety-like behavior. Brain Res. 1044, 116—121. Tannure, R.M., Bittencourt, A.S., Schenberg, L.C., 2009. Short-term full kindling of the amygdala dissociates natural and periaqueductal gray-evoked flight behaviors of the rat. Behav. Brain Res. 199, 247—256. Tye, K.M., Prakash, R., Kim, S.Y., Fenno, L.E., Grosenick, L., Zarabi, H., Thompson, K.R., Gradinaru, V., Ramakrishnan, C., Deisseroth, K., 2011. Amygdala circuitry mediating reversible and bidirectional control of anxiety. Nature 471, 358—362. Vidal-Gonzalez, I., Vidal-Gonzalez, B., Rauch, S.L., Quirk, G.J., 2006. Microstimulation reveals opposing influences of prelimbic and infralimbic cortex on the expression of conditioned fear. Learn. Mem. 13, 728—733. Wall, V.L., Fischer, E.K., Bland, S.T., 2012. Isolation rearing attenuates social interaction-induced expression of immediate early gene protein products in the medial prefrontal cortex of male and female rats. Physiol. Behav. 107, 440—450. Wallace, T.L., Stellitano, K.E., Neve, R.L., Duman, R.S., 2004. Effects of cyclic adenosine monophosphate response element binding protein overexpression in the basolateral amygdala on behavioral models of depression and anxiety. Biol. Psychiatry 56, 151—160. Yehuda, R., Teicher, M.H., Trestman, R.L., Levengood, R.A., Siever, L.J., 1996. Cortisol regulation in posttraumatic stress disorder and major depression: a chronobiological analysis. Biol. Psychiatry 40, 79—88.